348 



RAYMOND PEARL 



TABLE II!. 



SHOWING THE CLASSIFICATION OF THE KERNELS OF EAR No. 9 BY THE DIFFERENT 



OBSERVERS. 



1. For no one of the ears is there entire agreement among 

 all the observers as to the number of kernels falling in any one 

 of the color classes. There is entire agreement among the ob- 

 servers as to the total number of starchy and sweet kernels in 

 the case of two ears (Nos. 10 and 1 1), leaving out of account the 

 loss of one starchy kernel from ear No. 10 between the time 

 when this ear was counted by observers X. and XI. In the 

 case of the other two ears (Nos. 8 and 9) there is some disagree- 

 ment as to the number of starchy and sweet kernels. In no 

 case, however, is the disagreement in regard to these characters 

 so marked as that in respect to color characters. 



2. The relative amount of divergence among the observers in 

 regard to the distribution of the kernels in color classes is strik- 

 ingly different for different cars. Ear No. 9 plainly bore kernels 

 which were relatively easy to classify. The same was true of 

 ear No. 10. On the other hand the kernels of ears 8 and II 

 offered many difficulties in classification. But in the case of 

 ear No. i i the difficulty was largely confined to the sweet kernel-, 

 there being close agreement between all the observers but one 



